BRIEF REPORT

Animal-Assisted Therapy for Elderly
Schizophrenic Patients
A One-Year Controlled Trial
Yoram Barak, M.D., Osnat Savorai, B.A.
Svetlana Mavashev, B.A., Avshalom Beni, B.A.

Animal-assisted therapy (AAT) has been used as a therapeutic tool in various psychiatric populations, but there have been no published studies with elderly schizophrenic
patients. The authors evaluated, in a blinded, controlled manner, the effects of AAT in
a closed psychogeriatric ward over 12 months. Subjects were 10 elderly schizophrenic
patients and 10 matched patients (mean age: 79.1‫ 7.6ע‬years). The outcome measure
was the Scale for Social Adaptive Functioning Evaluation (SAFE). AAT was conducted
in weekly 4-hour sessions. Treatment encouraged mobility, interpersonal contact, and
communication and reinforced activities of daily living (ADLs), including personal
hygiene and independent self-care, through the use of cats and dogs as “modeling
companions.” The SAFE scores at termination showed significant improvement compared with baseline scores and were significantly more positive for the AAT group on
both Total SAFE score and on the Social Functions subscale. AAT proved a successful
tool for enhancing socialization, ADLs, and general well-being. (Am J Geriatr Psychiatry
2001; 9:439–442)

ince Gunby’s1 observation that an elderly patient
who is progressing well may have a pet, more than
1,000 references have been published focusing on the
use of a variety of animals in the therapeutic setting.2
However, research to identify the mechanisms through
which animal-assisted therapy (AAT) exerts its influence
and to define target populations is sorely needed.3 The
scope of indications for AAT is wide, ranging from rehabilitation and hypertension4,5 to depression6 and Alzheimer disease.7 Just as dispersed are the theoretical
concepts cited as underlying AAT. The functions of animals within therapy are thought to be as a companion,

S

social facilitator, or substitute for close interpersonal relationships. Animals are also thought to enhance health
status, increase sensory stimulation, provide emotional
support, reinforce feelings of independence, and facilitate behavior modification.8–10
Whatever the theoretical background, there are accumulating data that AAT modifies social behavior between two or more people.11 In a recent study of nearly
1,000 elderly community-dwelling subjects followed for
1-year, pet ownership contributed to maintenance of
enhanced activities of daily living (ADL) levels.12
Schizophrenia is one of the disorders in which un-

Received July 20, 2000; revised December 7, 2000; accepted January 22, 2001. From Abarbanel Mental Health Center, Bat Yam, Israel. Address
correspondence to Dr. Barak, Director, Psychogeriatric Dept., Abarbanel Mental Health Center, Bat-Yam, 59100, ISRAEL. e-mail:
mdybarak@netvision.net.il
Copyright ᭧ 2001 American Association for Geriatric Psychiatry

Am J Geriatr Psychiatry 9:4, Fall 2001

439

Animal-Assisted Therapy
favorable outcome is associated with a deficit in social
functioning.13 Coupled with the reduction in communication and social functioning that is reported in aging,14 geriatric schizophrenic patients are a subgroup
that suffers from severe impairment in social functioning. An additional variable adversely affecting this population is long-term hospitalization in state hospitals
wherein stimuli and demands for social interaction are
minimal.
The aim of the present study was to evaluate the
effects of AAT on long-stay geriatric schizophrenic patients in a controlled 1-year study.

METHODS
Subjects
Subjects in the study were 20 chronic schizophrenic patients, who were long-stay residents at the
Abarbanel Mental Health Center, Bat Yam, Israel. The
center is a tertiary-care, university-affiliated hospital
providing services to an urban catchment area of
800,000 patients. The psychogeriatric division is composed of three wards, with 34 beds each. One of the
wards, (in which this study was conducted ), provides
services to elderly patients suffering from schizophrenia. Patients are all long-stay residents (mean duration
of current hospitalization: 27‫ 1.7ע‬years), very similar
to the “very-difficult-to-place” elderly schizophrenia patients retained in state hospitals in the United States, as
described by White et al.15
Inclusion criteria for participation were the following: 1) DSM-IV diagnosis of schizophrenia according to
the Structured Clinical Interview (SCID–Hebrew Version)16; 2) age: Ն65 years; 3) current hospitalization:
Ն10 years; and 4) informed consent. Informed consent
was obtained after detailed explanation of the study
aims, after approval of this study by the hospital’s ethics
committee.
Exclusion criteria were the following: 1) severe
cognitive impairment; 2) allergic reactions to dogs or
cats; 3) physical illness exacerbated in the presence of
furry animals (e.g., asthma); and 4) planned discharge
within a 1-year period.
Patients were randomly selected for either the Animal Assisted Treatment group (AAT) or the Control
group. Average age was 79.1‫ 4.7ע‬years. In both
groups, there were seven women and three men.

440

Assessment
The primary outcome measure for the present
study was a change in the Social-Adaptive Functioning
Evaluation (SAFE) scores. The SAFE was developed by
Harvey et al.17 with established psychometric and predictive properties. The instrument consists of 17 items,
each rated on a 5-point scale (0‫ס‬no impairment;
4‫ס‬maximal impairment). The items in the scale measure social–interpersonal, instrumental, and life-skills
functioning and are designed to be rated by observation
and interaction with the subject. Patients were rated on
the SAFE before the study, after 6 months of treatment,
and upon completion (at 12 months).
Ratings were conducted by a clinical psychologist
blinded to the patients’ status (i.e., treatment [AAT] or
control).

Procedure
AAT was undertaken once weekly on the same day.
The therapists and assisting animals came to the ward
at 10:30 A.M., and the group session lasted 3 hours.
Three AAT counselors from the PET (Pet Enrichment
Therapy) program at Kibbutz Givat Haim Ichud, Israel,
were regularly accompanied by a psychiatric nurse, providing a ratio of 1:2.5 caretakers to patients. Each patient was provided with his own dog or cat, according
to personal preference.
Sessions included “ADL modeling activities” such as
petting, feeding, grooming, bathing, and teaching the
animals to walk on a lead for greater mobility.
Another major component of the treatment plan
was to increase mobility and socialization through walking the animals outside the hospital grounds. These excursions facilitated interaction with people of all ages
outside, who were inevitably drawn to these very special animals. Each session was concluded in the ward
with a summation of the day’s activities and a special
time allocated for a “parting between friends”—between the patients, PET staff members, and their animal
assistants.
Control-group patients were assembled for reading
and discussion of current news for a similar duration on
the same days that AAT was undertaken. These sessions
were conducted by three certified nurses so as to keep
the ratio of staff to patients equal to that of the AAT
group.

Am J Geriatr Psychiatry 9:4, Fall 2001

Barak et al.
Analysis

ing and grooming, dressing, eating, neatness, and orientation) improved in both groups, but this change did
not attain statistical significance; and 3) social functioning improved significantly in the AAT group (P‫.)100.0ס‬
The social functioning dimension of the SAFE scale consists of the following items: conversational skills, instrumental social skills, social appropriateness/politeness,
social engagement, friendships, recreation/leisure, communication skills, and participation in hospital programs.

Statistical analyses were performed by a medical
statistician of Technostat Ltd., Israel. Within- and between-groups comparisons were tested by the paired/
unpaired Student t-test (respectively). Where the assumption of normality was not met by distribution of
scores, the Wilcoxon nonparametric test was performed. All tests were two-tailed, and P values Յ0.05
were considered statistically significant. Demographic
and clinical variables are presented as mean‫ע‬standard
deviation (SD).

DISCUSSION
RESULTS

The present study demonstrates a significantly positive
effect of AAT for elderly schizophrenic patients in longterm settings. The major effect noted was in the domain
of social functioning. This effect was apparent by 6
months of treatment and held true by the end of the
study, at 12 months. This is, to the best of our knowledge, the first controlled study of AAT in elderly schizophrenic patients. However, the encouraging results do
not address the complex issue of how this change was
brought about.
Fick, in 1993,18 demonstrated an increase in verbal
social interactions among nursing home residents in the
presence of a dog. Although, as early as 1980,9 a survey
of case histories, anecdotal evidence, and pilot studies
shows that pet animals facilitate rapport, no uniform
conceptualization of AAT has been offered to date. Our
patients were initially ambivalent toward the AAT program. The program enabled the patients to view animals, people, and inanimate objects, as one patient said,
“like they once were.” Thus, the reintroduction of formerly stressful images was undertaken, with emphasis
on group interaction. Patients were encouraged to share
feelings of isolation and loneliness by first confiding in
their animal companions and later with their group co-

All subjects completed the study. During the study, one
patient in the AAT group and two patients in the Control group were hospitalized in a general-medical hospital because of intercurrent illness. All three patients
returned within 3 weeks.
Table 1 presents the SAFE scale scores preceding
the study (baseline), after 6 months, and at completion
(12 months). Total score and the three factors found in
analysis of the SAFE15 to reflect dimensions of adaptive
functioning in chronic schizophrenia patients are
shown in Table 1.
Statistical analysis (between-groups t-test) demonstrated that the AAT group’s SAFE total score improved
significantly as compared with the Control group. Improvement was already significant by 6 months
(P‫ )300.0ס‬and was maintained until the end of the
study (P‫.)100.0ס‬
Regarding the three factors of the SAFE, (withingroup tests): 1) impulse control (items include impulse
control, respect for property, money management, and
cooperation with treatment) did not change in either
group; 2) instrumental and self-care (items include bathTABLE 1.

SAFE Scale Total and factor scores throughout the study
AAT Group

Control Group

Baseline
Instrumental and Self-Care
Impulse Control
Social Functioning
Total

6 Months

12 Months

Baseline

6 Months

12 Months

31.1‫3.4ע‬

29.8‫8.3ע‬

27.5‫7.5ע‬

33.5‫1.6ע‬

29.1‫7.5ע‬

30.3‫5.5ע‬

8.2‫2.1ע‬

8.3‫7.1ע‬

7.9‫0.2ע‬

7.9‫1.2ע‬

7.9‫8.1ע‬

8.1‫6.1ע‬

24.5‫1.3ע‬

18.3‫**1.3ע‬

11.2‫**7.2ע‬

22.7‫4.3ע‬

24.1‫9.2ע‬

23.9‫1.3ע‬

42.6‫6.6ע‬

31.1‫*3.5ע‬

27.6‫**7.4ע‬

43.1‫4.6ע‬

39.2‫1.7ע‬

37.1‫4.5ע‬

Note: The SAFE Total score is not a summation of the subscale scores. Comparisons are between AAT group scores and the Control Group
scores. AAT‫ס‬animal-assisted therapy.*PϽ0.05; **PϽ0.01.

Am J Geriatr Psychiatry 9:4, Fall 2001

441

Animal-Assisted Therapy
members. The cats and dogs also served as “role models” for self-care and personal hygiene as they were
bathed, groomed, and fed by the patients, with emphasis on the differences among and unique needs of each
pet.
One of the major impediments to socialization in
schizophrenia is a deficit in adaptive life functioning.13,14 Because long-stay elderly schizophrenic patients spend many years in an impoverished environment, they may not have the opportunity to express
available social skills.19 The introduction of pets in a
supervised manner into this environment created a
novel “cue” for transference. However, AAT presents a

dynamic and constantly changing transitional object20
that reawakens both memories of a former “wholesome” life and a need to actively interact with the animals. This unique ability of AAT may be the mechanism
through which social functioning has improved in the
present study.
The use of AAT with elderly schizophrenic patients
needs further study. The limitations of our program are
the small sample size and relatively homogeneous population. However, the controlled design and quantified
outcome measure used encourage future use of AAT
with this needy group of patients.

References
1. Gunby P: Patient progressing well? he may well have a pet. JAMA
1979; 241:438
2. Draper RJ, Gerber GJ, Laying EM: Defining the role of pet animals
in psychotherapy. Psychiatric Journal of the University of Ottawa
1990; 15:169–172
3. Beck AM: The therapeutic use of animals. Vet Clin North Am
Small Anim Pract 1985; 15:365–375
4. Propsner N: Pets helping to heal the elderly. N Engl J Med 1997;
94:32
5. Black HR: Nonpharmacologic therapy for hypertension in the
elderly. Geriatrics 1989; 44(supplB):20–29
6. Cooper AM, Shamoian CA: Depression as seen by the psychiatrist. Dis Mon 1979; 25:1–64
7. Mimiran M, Van Gool WA, Van Haaren F, et al: Environmental
influences on brain and behavior in aging and Alzheimer’s disease. Prog Brain Res 1986; 70:443–459
8. Wilson CC, Netting FE: Companion animals and the elderly: a
state-of-the-art summary. J Am Vet Med Assoc 1983; 183:1425–
1429
9. Brickel CM: A review of the roles of pet animals in psychotherapy
and with the elderly. Int J Aging Hum Dev 1980; 12:119–128
10. Ryder EL: Pets and the elderly: a social work perspective. Vet
Clin North Am Small Anim Pract 1985; 15:333–343
11. Messent PR: Pets as social facilitators. Vet Clin North Am Small
Anim Pract 1985; 15:387–393
12. Raina P, Waltner-Toews D, Bonnett B, et al: Influence of compan-

442

ion animals on the physical and psychological health of older
people: an analysis of a one-year longitudinal study. J Am Geriatr
Soc 1999; 47:323–329
13. Perlick D, Stastny P, Mattis S, et al: Contributors of family, cognitive, and clinical dimensions to long-term outcome in schizophrenia. Schizophr Res 1992; 6:257–265
14. Kemper S: Life-span changes in synaptic complexity. J Gerontol
1987; 42:323–328
15. White L, Parrella M, McCrystal-Simon J, et al: Characteristics of
elderly psychiatric patients retained in a state hospital during
downsizing: a prospective study with replication. Int J Geriatr
Psychiatry 1997; 12:474–480
16. DSM-IV diagnosis of schizophrenia according to the Structured
Clinical Interview: SCID–Hebrew Version. Hadassa University
Press, Jerusalem, 1995
17. Harvey PD, Davidson M, Mueser KT, et al: Social-Adaptive Functioning Evaluation (SAFE): a rating scale for geriatric psychiatric
patients. Schizophr Bull 1997; 23:456–463
18. Fick KM: The influence of an animal on social interactions of
nursing home residents in a group setting. Am J Occup Ther
1993; 47:529–534
19. Tessler RC, Manderscheid RW: Factors affecting adjustment to
community living. Hosp Community Psychiatry 1982; 33:203–
207
20. Beck AM, Katcher AH: A new look at pet-facilitated therapy. J
Am Vet Med Assoc 1985; 184:414–421

Am J Geriatr Psychiatry 9:4, Fall 2001